Natural environment and gestational diabetes risk in Australia: a spatiotemporal ecological regression approach.

BACKGROUND

Disproportionate exposure to natural environment attributes (e.g., greenness, air quality) may be associated with disparities in gestational diabetes mellitus (GDM) risk across areas and social factors. We examined the association between the natural environment features and GDM risk, and explored variation by area-level social factors.

METHODS

A nationwide ecological spatiotemporal study at Statistical Area Level 2 (SA2), a medium-sized spatial resolution, was conducted using data from 2016 to 2022. SA2-level annual data on GDM cases, births, environmental exposures (air pollution [PM2.5 and NO2], greenness, and temperature), and neighbourhood social factors (e.g., socioeconomic status) were used. Residential greenness was measured using Normalised Differential Vegetation Index (NDVI). A spatiotemporal ecological regression approach supported by a Bayesian framework was applied. Effect modification analysis was conducted to explore whether the association between environmental exposures and the risk of GDM varied across different neigbourhood-level social factors.

RESULTS

We included 241,264 GDM cases among 2,035,100 women across 1,977 SA2s from 2016 to 2022. An 11% (Adjusted Risk Ratio [ARR]: 0.89 [0.83-0.95]) reduction in GDM risk was associated with an increase in residential greenness (per 0.10 increase in NDVI). This association was stronger in areas of most socioeconomic advantage (ARR: 0.70 [0.59-0.83]) and in neighbourhoods with a high concentration of non-European migrant women (ARR: 0.83 [0.76-0.89]). GDM risk was associated with high PM2.5 levels (> 5 µg/m3), with a 43% higher risk in areas of most socioeconomic disadvantage (ARR: 1.43 [1.11-1.88]) and a 23% higher risk in areas with high concentrations of non-European migrant women (ARR: 1.23 [1.05-1.45]). There was no significant association between NO2 and GDM risk, ARR: 1.01(95% CrI 1.00, 1.95]).

CONCLUSIONS

High residential greenness may be associated with a lower risk of GDM, with potential differences by social factors. The increased GDM risk associated with high PM2.5 was more pronounced in areas of socioeconomic disadvantage and in areas with high concentrations of non-European migrant women. NO2 did not show a significant association with GDM risk. These findings suggest that geographically targeted interventions may help mitigate the risk of GDM associated with environmental exposures, particularly among vulnerable populations.